ihsaan-ullah/academic-papers-dataset

📚 Academic Papers Dataset

This dataset accompanies the paper:

“Evaluating LLM Judgment via Surgical Edits”

GitHub Repository:
🔗 https://github.com/ktgiahieu/eval_llm_surgical_edits

---

📝 Dataset Summary

The Academic Papers Dataset contains structured, parsed scientific papers from:

• ICLR → 2024, 2025
• ICML → 2024, 2025
• NeurIPS → 2021, 2022, 2023, 2024

The dataset provides each paper in a standardized Markdown format, along with extracted figures.

---

📊 Dataset Statistics

Conference Paper Counts

Conference	Accepted (Submitted)	Accepted (Parsed)	Rejected (Submitted)	Rejected (Parsed)
ICLR 2024	2260	949	709	3438
ICLR 2025	3704	1406	1078	4922
ICML 2024	2944	1065	0	N/A
ICML 2025	3260	1141	42	N/A
NeurIPS 2021	2334	1116	46	133
NeurIPS 2022	2671	1132	47	153
NeurIPS 2023	3218	1205	0	176
NeurIPS 2024	4035	1574	0	201

Total accepted (submitted): 24,426
Total accepted (parsed): 8,588
Total rejected (submitted): 1,922
Total rejected (parsed): 9,023

Note: The paper submission counts for each conference were obtained using the publicly available statistics from PaperCopilot (https://papercopilot.com/).

---

🌍 Languages

• English (en)
  (Official conference submissions from ICLR, ICML, NeurIPS)

---

🚀 How to Use This Dataset

Because the dataset contains large ZIP archives and thousands of structured papers, we recommend using git-xet for efficient cloning and storage.

1️⃣ Install git-xet

git-xet enables fast and storage-efficient checkout of large files from Hugging Face Datasets.

Official docs: https://hf.co/docs/hub/git-xet

If you're on macOS:

brew install git-xet
git xet install

Linux:

curl https://gitxet.com/install.sh | sudo bash
git xet install

2️⃣ Clone the dataset repository

git clone https://huggingface.co/datasets/ihsaan-ullah/academic-papers-dataset

This will download lightweight file pointers first, and git-xet will fetch the large paper archives on demand.

📂 Dataset Structure

Inside the cloned repo, you will find per-conference ZIP files:

To extract:

unzip {ConferenceName}.zip -d {ConferenceName}/

After extraction, you will see:

content/
└── parsed_arxiv/
    └── {ConferenceName}/
        └── accepted/
            └── structured_paper_output/
                ├── paper.md
                ├── figures/
                │   ├── fig1.png
                │   ├── fig2.png
                │   └── ...
        └── rejected/
            └── structured_paper_output/
                ├── paper.md
                ├── figures/
                │   ├── fig1.png
                │   ├── fig2.png
                │   └── ...

File Descriptions

File	Purpose
paper.md	Markdown representation of the full scientific paper
figures/	Extracted figures with their captions

---

📜 Citation

TBA